^^ there is a mathematical explanation in the GHOST paper. See figure 1 for actual data on block propagation times with varying block sizes.
If anyone is insisting you need >50% of the network hash rate to double spend, they're incorrect. That's simply the quantity of hash rate required so that on average you are able to double spend 100% of the time. You can do it with a lesser hash rate, with lower probability.
Because it's possible doesn't mean it'll happen -- deepbit and litecoinpool were often > 51% and attacks never occurred. But when you have a poorly secured new coin...
Topic: Importance of Speed (Read 1122 times)
Higher block speed also puts you at unique security disadvantages, as propagation time amongst all nodes becomes an issue. The larger the network and blocksize, the greater this becomes an issue.
Basically, the tradeoff is like this: In PoW, blocks contain many tx. To effectively be a solution for the Two General's Problem, each node relaying a new block must download and verify its contents completely before doing so. However, a malicious attacker instead begins generating a bunch of his own blocks that are not revealed to the network. Because the miner does not have to rely on his blocks being relayed throughout the network, he is given an advantage in terms of being able to perform a 51% attack and can now do so with less than 51%; in the case of very fast blocks with large block sizes, this is probably way less than 51%.
People will likely find this out as soon as an exchange loses a lot of money on a double spend; then you need to crank the block size down, and end up overall with not much of an advantage compared to Bitcoin when you're considered tx/sec, and you'll likely have to spend more in fees just to get your tx confirmed quickly because the supply of tx in a block is greatly diminished while demand runs the same.
Anyway, don't say I didn't warn you...
Basically, the tradeoff is like this: In PoW, blocks contain many tx. To effectively be a solution for the Two General's Problem, each node relaying a new block must download and verify its contents completely before doing so. However, a malicious attacker instead begins generating a bunch of his own blocks that are not revealed to the network. Because the miner does not have to rely on his blocks being relayed throughout the network, he is given an advantage in terms of being able to perform a 51% attack and can now do so with less than 51%; in the case of very fast blocks with large block sizes, this is probably way less than 51%.
People will likely find this out as soon as an exchange loses a lot of money on a double spend; then you need to crank the block size down, and end up overall with not much of an advantage compared to Bitcoin when you're considered tx/sec, and you'll likely have to spend more in fees just to get your tx confirmed quickly because the supply of tx in a block is greatly diminished while demand runs the same.
Anyway, don't say I didn't warn you...
I did a fair bit of research, and spoke to some creator of coins.
and I want to counter argue that the above scenario is extremely unlikely/impossible, and the only way to do it, is to have a 51% attack.
Also point being, if it is possible, it would have already happened, given the huge volume daily and profit bounty.
I would like to ask Tachotime, why do you even think it is possible? I want to learn what you see.
you are right, currently people will take more confirmation for quicker coins, resulting in overall similar time.
I am thinking when the next step comes, and when people will to accept after 2-3 confirmations. Given a high network hashrate, security shouldn't be too high of an issue, and with faster coins, the initial confirmation will come sooner. that's what I am thinking about.
do you think I am considering too few parameters?
you are right, currently people will take more confirmation for quicker coins, resulting in overall similar time.
I am thinking when the next step comes, and when people will to accept after 2-3 confirmations. Given a high network hashrate, security shouldn't be too high of an issue, and with faster coins, the initial confirmation will come sooner. that's what I am thinking about.
do you think I am considering too few parameters?
Higher block speed also puts you at unique security disadvantages, as propagation time amongst all nodes becomes an issue. The larger the network and blocksize, the great this becomes an issue.
Basically, the tradeoff is like this: In PoW, blocks contain many tx. To effectively be a solution for the Two General's Problem, each node relaying a new block must download and verify its contents completely before doing so. However, a malicious attacker instead begins generating a bunch of his own blocks that are not revealed to the network. Because the miner does not have to rely on his blocks being relayed throughout the network, he is given an advantage in terms of being able to perform a 51% attack and can now do so with less than 51%; in the case of very fast blocks with large block sizes, this is probably way less than 51%.
People will probably find this out as soon as an exchange loses a lot of money on a double spend; then you need to crank the block size down, and end up overall with not much of an advantage compared to Bitcoin when you're considered tx/sec, and you'll likely have to spend more in fees just to get your tx confirmed quickly because the supply of tx in a block is greatly diminished while demand runs the same.
Basically, the tradeoff is like this: In PoW, blocks contain many tx. To effectively be a solution for the Two General's Problem, each node relaying a new block must download and verify its contents completely before doing so. However, a malicious attacker instead begins generating a bunch of his own blocks that are not revealed to the network. Because the miner does not have to rely on his blocks being relayed throughout the network, he is given an advantage in terms of being able to perform a 51% attack and can now do so with less than 51%; in the case of very fast blocks with large block sizes, this is probably way less than 51%.
People will probably find this out as soon as an exchange loses a lot of money on a double spend; then you need to crank the block size down, and end up overall with not much of an advantage compared to Bitcoin when you're considered tx/sec, and you'll likely have to spend more in fees just to get your tx confirmed quickly because the supply of tx in a block is greatly diminished while demand runs the same.
This is the only person in this thread that has any idea what they're talking about. If cryptocurrencies could be magically faster with no drawbacks, do you guys not think that Bitcoin would have done it?
Higher block speed also puts you at unique security disadvantages, as propagation time amongst all nodes becomes an issue. The larger the network and blocksize, the greater this becomes an issue.
Basically, the tradeoff is like this: In PoW, blocks contain many tx. To effectively be a solution for the Two General's Problem, each node relaying a new block must download and verify its contents completely before doing so. However, a malicious attacker instead begins generating a bunch of his own blocks that are not revealed to the network. Because the miner does not have to rely on his blocks being relayed throughout the network, he is given an advantage in terms of being able to perform a 51% attack and can now do so with less than 51%; in the case of very fast blocks with large block sizes, this is probably way less than 51%.
People will likely find this out as soon as an exchange loses a lot of money on a double spend; then you need to crank the block size down, and end up overall with not much of an advantage compared to Bitcoin when you're considered tx/sec, and you'll likely have to spend more in fees just to get your tx confirmed quickly because the supply of tx in a block is greatly diminished while demand runs the same.
Anyway, don't say I didn't warn you...
Basically, the tradeoff is like this: In PoW, blocks contain many tx. To effectively be a solution for the Two General's Problem, each node relaying a new block must download and verify its contents completely before doing so. However, a malicious attacker instead begins generating a bunch of his own blocks that are not revealed to the network. Because the miner does not have to rely on his blocks being relayed throughout the network, he is given an advantage in terms of being able to perform a 51% attack and can now do so with less than 51%; in the case of very fast blocks with large block sizes, this is probably way less than 51%.
People will likely find this out as soon as an exchange loses a lot of money on a double spend; then you need to crank the block size down, and end up overall with not much of an advantage compared to Bitcoin when you're considered tx/sec, and you'll likely have to spend more in fees just to get your tx confirmed quickly because the supply of tx in a block is greatly diminished while demand runs the same.
Anyway, don't say I didn't warn you...
Here are the following category for speed, where do you think is the level you think block speed should be at?
Question doesn't make sense without comparing the corresponding levels of security. 15 second confirmation would be 15/(60*10)= 0.025 as secure as a 10 minute confirmation _if_ the short confirmation time did not make the network less efficient. But we all know it does make the network less efficient so the real number is something much smaller... Where is the limit where you need to start waiting for multiple confirmations? Or if you don't, how is that substantially different from accepting unconfirmed small payments?In other words "Quark and Doge look good" does not have much evidence behind it yet...
I think bitcoin confirmation time is good when we are doing online buying, etc.
i am thinking about.. what if the service is expanded to retail, like supermarket, with online wallets, etc.
that being the interface, I feel we need something faster than 10min per person waiting in line.
in that aspect, Dogecoin/Worldcoin/Quark has a unique advantage.
The confirmation time of bitcoin is more than short enough for me.
It seldom happens that I need to make a payment that needs to have x confirmations within the hour or so.
It seldom happens that I need to make a payment that needs to have x confirmations within the hour or so.
http://www.quarkcoin.cc/2013/12/14/mythbuster-pre-mine-with-max-g-qrk-main-developer/
Quark was NOT mined, get educated.
Quark was NOT mined, get educated.
You should probably use coins that is actual on equal playing fields and have actual info displayed with source.
http://bitinfocharts.com/
Anything after 30 second will cause too many orphan blocks.
http://bitinfocharts.com/
Anything after 30 second will cause too many orphan blocks.
Wow! Such chart!
Quark and Doge look good
while we can probably all agree that bitcoin block time is too slow at 20 mins. (you may disagree)
I disagree, when it's 10mins in reality., besides, if you've ever bought something with bitcoin through a payment processor, it's instant.
K.
on the faster coins, I feel there's something worth talking about orphan blocks.
according to http://bitinfocharts.com/
quark has 6.7% orphan blocks (kinda high)
I will say any orphan block percentage over 3% really is too high.
anyone agree with me there?
Fastcoin is the best in this way.
quark is instamined not premined and dogecoin is a clonecoin .
You can't really call it instamined. They just used a new economical model that help create stability in the coin. It is definitely an unique point of Quark.
quark is instamined not premined and dogecoin is a clonecoin .
Worldcoin doesn't have many orphan blocks for 30 secs yet.
but you are right, fast coin is riddled with orphans.
hmmmm. interesting, I post this, cause personally I think speed is crucial, and as such I decided to see what everyone thinks.
but you are right, fast coin is riddled with orphans.
hmmmm. interesting, I post this, cause personally I think speed is crucial, and as such I decided to see what everyone thinks.
Speed is definitely important factor. Speed, security, and stability the three top thing should be considered in a coin. Network security is also important.
Worldcoin doesn't have many orphan blocks for 30 secs yet.
but you are right, fast coin is riddled with orphans.
hmmmm. interesting, I post this, cause personally I think speed is crucial, and as such I decided to see what everyone thinks.
but you are right, fast coin is riddled with orphans.
hmmmm. interesting, I post this, cause personally I think speed is crucial, and as such I decided to see what everyone thinks.
Yep everyone buy quark this premined coin that was only going high because of max keiser.
Oh i see he has his own coin now.. well..
Oh i see he has his own coin now.. well..
Quark was never "pre-mined" Quarkcoin First Block found on 2013-07-21 and announced on 2013-07-21.
Source:
http://bitinfocharts.com/quarkcoin/
https://bitcointalksearch.org/topic/ann-qrk-quark-core-010-upgrade-260031
nice info
Jump to: